CENTER FOR DRUG EVALUATION AND RESEARCH
`
`
`
`
`
`APPLICATION NUMBER:
`22511Orig1000
`
`
`
`SUMMARY REVIEW
`
`
`
`
`
`

`

`Summary Review for Regulatory Action
`
`Division Director Review
`
`
`
`
`Date
`From
`Subject
`NDA#
`Applicant Name
`Date of Submission
`
`(electronic stamp)
`Donna Griebel, MD
`Division Director Summary Review
`022511
`Pozen Inc.
`June 30, 2009
`Received: July 7, 2009
`April 30, 2010
`Vimovo
`naproxen and esomeprazole magnesium
`Naproxen and esomeprazole magnesium tablets:
`375 mg naproxen/20 mg esomeprazole magnesium
`AND
`500 mg naproxen/ 20 mg esomeprazole magnesium
`Treatment of signs and symptoms of osteoarthritis,
`rheumatoid arthritis, and ankylosing spondylitis in patients
`at risk for developing NSAID-associated gastric ulcers
`Approval
`
`PDUFA Goal Date
`Proprietary Name /
`Established (USAN) Name
`Dosage Forms / Strength
`
`Proposed Indication(s)
`
`Action:
`
`
`Material Reviewed/Consulted
`OND Action Package, including:
`Medical Officer Review
`
`Biostatistical Review
`
`Pharmacology Toxicology Review
`CMC Review
`Clinical Pharmacology Review
`Biopharmaceutics Review
`DDMAC
`DSI
`
`CDTL Review
`OSE/DMEPA
`
`SEALD
`PMHS
`
`
`Names of discipline reviewers
`Erica Wynn, MD/ Ruyi He, MD/Jin Chen, MD/Ellen
`Fields, MD
`Freda Cooner, PhD/Kate Meeker, MS/Mike Welch,
`PhD
`Sushanta Chakder, PhD
`Rajiv Agarwal/ Moo Jhong Rhee, PhD
`Jane Bai, PhD/Dilara Jappar, PhD/Sue-Chih Lee, PhD
`Tien-Mien Chen, PhD/Patrick Marroum PhD
`Katie Klemm/Lisa Hubbard/Shefali Doshi/Robert Dean
`Sripal Mada, PhD/Sean Kassim, Ph.D./C.T.
`Viswanathan, PhD
`Ruyi He, MD
`Kellie Taylor, PharmD, MPH/Denise Toyer,
`PharmD/Carol Holquist, RPh/Zachary Oleszczuk,
`Pharm D
`Debbie Beitzell, BSN
`Alyson Karesh, MD/Hari Cheryl Sachs, MD/Lisa
`Mathis, MD
`
`OND=Office of New Drugs
`DDMAC=Division of Drug Marketing, Advertising and Communication
`OSE= Office of Surveillance and Epidemiology
`DMEPA=Division of Medication Error Prevention and Analysis
`
`Page 1 of 25
`
`

`

`Division Director Review
`
`DSI=Division of Scientific Investigations
`CDTL=Cross-Discipline Team Leader
`PMHS = Pediatric and Maternal Health Staff
`
`Page 2 of 25
`
`

`

`Division Director Review
`
`Division Director Review
`
`
`
`1. Introduction
`This NDA, submitted under 505(b)(2) section of the Federal Food, Drug and Cosmetic Act
`and 21 CFR Part 314.50, seeks approval of two dosage strengths of a fixed combination of
`naproxen and immediate release esomeprazole. The inner enteric coated core of the tablet is
`one of two strengths of naproxen, either 375 mg or 500 mg. The tablet in both naproxen
`dosage strengths is coated in an outer immediate release film that contains 20 mg of
`esomeprazole. Two approved NDAs were referenced: NDA 020067 for EC-Naprosyn (375
`mg and 500 mg) and NDA 021153 for Nexium capsules (20 mg and 40 mg). EC-Naprosyn is
`a delayed release naproxen tablet formulation approved for treatment of signs and symptoms
`of osteoarthritis, rheumatoid arthritis, and ankylosing spondylitis, at both dose strengths and
`dosed twice daily. Esomeprazole is approved for reducing the risk of developing NSAID-
`associated gastric ulcers, at doses of 20 mg and 40 mg, dosed once daily. The proposed dosing
`schedule for Vimovo is twice daily dosing.
`
`The Applicant proposes the following indication for Vimovo:
`
`VIMOVO is indicated for the relief of signs and symptoms of osteoarthritis, rheumatoid
`arthritis and ankylosing spondylitis in patients at risk of developing NSAID-associated gastric
`ulcers.
`
` VIMOVO is not recommended for initial treatment
`of acute pain because the absorption of naproxen is delayed compared to absorption from
`other naproxen containing products.
`
`This indication does not clearly state the role/indication for the esomeprazole component of
`Vimovo and does not make clear that there are two component products, each with a separate
`indication. In addition, there are two deviations in the proposed indication from the reference
`drug Nexium:
`
`
`The Applicant conducted pharmacokinetic/bioavailability studies of each component of
`Vimovo (naproxen 375 mg, naproxen 500 mg and esomeprazole 20 mg) to establish a bridge
`between Vimovo and the two referenced NDAs. Bioequivalence was established for the two
`naproxen doses; however, the immediate release esomeprazole component of Vimovo was not
`bioequivalent to the referenced Nexium product. This was anticipated, due to the immediate
`release formulation of esomeprazole in Vimovo, which makes it subject to degradation by
`gastric acid.
`
`The Applicant investigated the efficacy and safety of Vimovo for reduction of gastric ulcer in
`two phase 3 studies of 6 months duration in which the Vimovo 500 mg naproxen dosage form
`
`Page 3 of 25
`
`(b) (4)
`
`(b) (4)
`
`

`

`Division Director Review
`
`(500 mg naproxen/immediate release esomeprazole 20 mg) was compared to EC-Naprasyn
`500mg, both administered twice daily. Two additional phase 3 trials (active and placebo
`controlled) were conducted to evaluate the efficacy of the naproxen component of Vimovo for
`treatment of signs and symptoms of osteoarthritis.
`
`The development program was a collaborative effort of two companies, Pozen and Astra
`Zeneca. Under a licensing agreement between the companies, the NDA will be transferred to
`Astra Zeneca upon approval.
`
`No review issues preclude approval.
`
`
`2. Background
`Esomeprazole is a proton pump inhibitor that inhibits the H+/K+ ATPase enzyme system at
`the secretory surface of the gastric parietal cell. Esomeprazole is acid labile and is degraded
`by gastric acid to a cationic sulfonamide. The esomeprazole in Vimovo is an “immediate-
`release” formulation that exposes it to some degradation by gastric acid.
`
`Naproxen is a marketed nonsteroidal anti-inflammatory drug. One of the marketed
`formulations of naproxen is enteric coated, EC-Naprosyn. EC-Naprosyn is approved for the
`indications of rheumatoid arthritis, osteoarthritis and ankylosing spondylitis at both 375 mg
`and 500 mg doses, with twice daily administration. Esomeprazole is approved for risk
`reduction of NSAID-associated gastric ulcer at doses of 20 mg or 40 mg once daily for up to 6
`months (controlled studies do not extend beyond 6 months).
`
`
`3. CMC
`
` I
`
` concur with the conclusions reached by the chemistry reviewers that the NDA has provided
`sufficient CMC information to assure the identity, strength, purity and quality of the drug
`product. An “acceptable” recommendation was received from the Office of Compliance on
`March 24, 2010.
`
`The drug substance esomeprazole magnesium is manufactured in France by AstraZeneca
`Dunkerque Production. The naproxen drug substance is manufactured in
`
` The drug product, stability testing, bulk packaging and quality
`control testing is performed at Patheon Pharmaceuticals, Inc., in Cincinnati, Ohio. AstraZeneca
`Pharmaceuticals LP in Newark, Deleware performs packaging, labeling, quality control and
`batch release of drug product.
`
`The drug product, Vimovo, was designed as a fixed combination tablet of two distinct
`formulations. The inner enteric coated (delayed release) component of naproxen contains
`either 375 mg or 500 mg of naproxen and the outer immediate release film coat of
`esomeprazole magnesium contains 20 mg of esomeprazole (present as 22.3 mg of
`esomeprazole magnesium). The CMC reviewer noted that “based on the qualitative and
`
`Page 4 of 25
`
`(b) (4)
`
`

`

`Division Director Review
`
`quantitative formulation, the two strengths of [Vimovo] are dose proportional.” Formulation
`changes made after the phase 3 batches were considered minor.
`
`Biopharmaceutics
`The Biopharmaceutics reviewer identified no approvability issues. He determined that the
`Applicant’s proposed dissolution methodology should only be used on an interim basis and
`that the Applicant should submit, as a post-marketing commitment, additional dissolution
`testing data on the naproxen component of the tablets using the USP dissolution methodology
`for enteric coated drug products. The reason that the Applicant’s dissolution methodology for
`naproxen was not considered optimal was that it lacked a pre-exposure in acid stage to test the
`enteric coating of naproxen. In addition, the reviewers recommended that the Applicant
`tighten its dissolution specifications for the esomeprazole in the buffer stage.
`
`On February 24, 2010 FDA Biopharmaceutics review team requested a phase 4 commitment
`that the Applicant would implement by one year post-approval, transition from the proposed
`naproxen dissolution method in the NDA to the USP dissolution method for the naproxen
`component of the tablets. (See PMC in Approval Letter and in Section 13 of this review.) This
`agreement was documented in an amendment to the NDA on March 4, 2010. In an April 21,
`2010 teleconference the FDA biopharmaceutics review team clarified that with reference to the
`post marketing commitment agreed to on March 4, the Applicant should generate new
`dissolution profile data utilizing the USP dissolution method and submit the data generated for
`both naproxen and esomeprazole. The FDA will reassess the interim specifications and
`consider if the new data generated are supportive of the interim specifications or if the data
`support revised specifications. The specifications selected and supported by data will then be
`adopted as the final specifications for naproxen and esomeprazole.
`
`
`4. Nonclinical Pharmacology/Toxicology
`This is a 505(b)(2) application. The Applicant submitted a nonclinical pharmacokinetic study
`in which urinary and plasma metabolites of buffered and unbuffered omeprazole were
`determined in rats following 14 days of oral dosing. The metabolite profiles for omeprazole
`were similar for the two formulations. Dr. Chakder agreed that oral administration of the
`uncoated esomeprazole in Vimovo should not lead to exposure of humans to new metabolites.
`I concur with the conclusions reached by the pharmacology reviewer, Dr. Chakder, that there
`are no outstanding pharmacology/toxicology issues that preclude approval. I concur with his
`labeling recommendations.
`5. Clinical Pharmacology
`
` I
`
` concur with the conclusions reached by the clinical pharmacology reviewers that there are no
`outstanding clinical pharmacology and biopharmaceutics issues that preclude approval. Their
`labeling recommendations were incorporated in label negotiations. The Clinical
`Pharmacology reviewers recommended the Applicant should be required to conduct
`pharmacokinetic studies in a pediatric population ages 2 years to 17 years.
`
`
`Page 5 of 25
`
`

`

`Division Director Review
`
`The Vimovo 500/20 mg tablet batches were tested in Phase 1 clinical pharmacology studies
`and in the clinical phase 3 pivotal studies. The lower naproxen dose tablet, Vimovo 375/20
`mg, was not tested in Phase 3 clinical trials; however, a primary stability batch of the 375/20
`mg tablet was tested in a phase 1 clinical pharmacology study. The reviewers determined that
`no biowaiver request was needed for this NDA. The excipients used in the two dose
`formulations of Vimovo are approximately dose proportional.
`
`The clinical pharmacology reviewers found that both doses of the referenced naproxen
`product, EC-Naprosyn, were bioequivalent to their respective Vimovo dose levels. The
`reviewers determined that the AUC of the 20 mg immediate release esomeprazole component
`of Vimovo after a single dose was approximately half that of the Nexium 20 mg reference.
`This was anticipated because the immediate release without gastric protection allows for
`degradation of the product by gastric acid. However, the Cmax was only slightly lower. The
`data from the single dose study are summarized in the table below, which is reproduced from
`Dr. Bai’s Clinical Pharmacology Review. (Vimovo is designated PN-400 in this table.)
`
`
`
`
`
`
`
`With repeat dosing (twice daily times 14 days) the AUC and Cmax were higher than after a
`single dose, especially with the morning dose. The PM dose was associated with lower
`esomeprazole concentrations than the morning dose. The esomeprazole repeat dose
`pharmacokinetic data for Vimovo (PN-400) are summarized in the table below, reproduced
`from Dr. Bai’s clinical Pharmacology review. High inter-and intraindividual variability in
`pharmacokinetics was noted for the esomeprazole component of Vimovo.
`
`
`
`
`
`
`Page 6 of 25
`
`

`

`Division Director Review
`
`
`
`
`
`The Day 14 AM Cmax is 3 times higher than the Day 1 AM Cmax. The Day 14 PM Cmax is
`1.9 times higher than the Day 1 PM Cmax. The AUC associated with the AM dose on Day 14
`is 4.7 times higher than the AM dose AUC on Day 1. The Day 14 PM dose AUC is 2.9 times
`higher than the Day PM AUC. This is briefly summarized below:
`
`
`
`
`
`
`
`
`Day 14 AM Cmax = 3x Day 1 Cmax
`Day 14 PM Cmax = 1.9 x Day 1 Cmax
`Day 14 AM AUC = 4.7 X Day 1 AUC
`Day 14 PM AUC = 2.9 X Day 1 AUC
`
`
`
`
`
`
`
`
`There are no head to head comparisons of multidose PK between the Vimovo esomeprazole
`and Nexium 20 mg. The Nexium NDA’s Clinical Pharmacology review includes a summary
`of two PK studies of 5-day multi-dosing (less than half of the duration of exposure in the
`Vimovo multi-dosing study). In these studies, increased exposure over 5 days was also
`documented. In a Nexium 40 mg study, the Cmax increased 2 fold and the AUC increased 2.6
`fold. In another study that utilized Nexium 20 mg, the AUC increased 1.8 fold.
`
`The Nexium NDA’s Clinical Pharmacology review also presents data from a study in which
`20 mg and 40 mg of esomeprazole were administered. The Cmax of the 40 mg dose was
`double that of the 20 mg dose level and the AUC was 3 fold higher than the 20 mg dose.
`Based on the single day dosing head to head comparison of Vimovo with Nexium presented
`above, and what is known about increased exposure over time with Nexium, the reviewers
`concluded that patients who take Vimovo will not be exposed to the Cmax achieved with a 40
`mg dose level of the approved Nexium product, and that the AUC would be similar to taking
`Nexium 20 mg BID, which would also not exceed the Nexium 40 mg once daily exposure.
`
` A
`
` pharmacodynamic study of varying doses of immediate release esomeprazole in the fixed
`combination of Vimovo was conducted, utilizing the endpoint of percent time of intragastric
`pH >4.0 on Day 9 of administration. The study included an enteric coated esomeprazole +
`naproxen arm. Based on this study, the to be marketed dose combination resulted in median of
`70% time with pH> 4.0 on Day 9, which exceeded the % time associated with the lower
`esomeprazole (10 mg ) combination and compared favorably within the study to the enteric
`coated esomeprazole + naproxen arm. Those data are summarized in the table below, which is
`reproduced from Dr. Bai’s Clinical Pharmacology review.
`
`
`Page 7 of 25
`
`

`

`Division Director Review
`
`
`
`The Clinical Reviewer expressed concerns that the Applicant did not study a 10 mg
`esomeprazole combination formulation in phase 3 efficacy trials. However, the Clinical
`Pharmacology reviewer states in her review that the 20 mg esomeprazole dose was selected
`based on the higher percentage of time with intragastric pH>4 associated with the 20 mg dose
`(71%) relative to the 10 mg esomeprazole dose (41%).
`
`No drug drug interaction between esomeprazole and naproxen was observed with
`coadministration. High fat meals reduced esomeprazole bioavailability by 50% and
`substantially delayed naproxen absorption (note shifts in Tmax in table below). Administration
`30 to 60 minute prior to a meal decreased the impact of food. The summary food effects PK
`for naproxen are summarized in the table below, which is reproduced from Dr. Bai’s review
`(PN400 = Vimovo).
`
`
`Page 8 of 25
`
`

`

`Division Director Review
`
`
`
`
`6. Clinical Microbiology
`Not applicable.
`7. Clinical-Efficacy
`
`
`
``
`Two identical phase 3 studies (Study 301 and 302) were submitted to support the efficacy of
`Vimovo in reducing the risk of developing NSAID-associated gastric ulcers (evaluation of the
`immediate release esomeprazole component of Vimovo). The 500 mg naproxen dose of the
`fixed combination Vimovo was studied in both studies. A third trial (Study 303) was initiated
`in a high risk population, but was terminated early due to insufficient enrollment (n=20). The
`Applicant also conducted two phase 3 trials, Study 307 and Study 309, to evaluate the efficacy
`of Vimovo (its naproxen component) for treating the signs and symptoms of osteoarthritis.
`Both studies enrolled patients with osteoarthritis of the knee and were designed to evaluate
`noninferiority to celecoxib. Reviewers from the Division of Gastroenterology Products (DGP)
`reviewed Studies 301 and 302, and reviewers from the Division of Anesthesia and Analgesia
`Products (DAAP) were consulted to review Studies 307 and 309. (DAARP had been
`previously consulted regarding the study design of these two trials at the post-end of Phase 2
`meeting in June 2008 and at the pre-NDA meetings.)
`
`Gastric Ulcer Risk Reduction Trials
`Studies 301 and 302 were identical randomized, double blind, parallel-group, multicenter
`clinical trials of 6 months duration. They were conducted in the United States. Patients were
`eligible for inclusion if they had a history of osteoarthritis, rheumatoid arthritis, ankylosing
`spondylitis, or other medical condition expected to require daily NSAID therapy for at least six
`months. In addition, there were specific eligibility criteria based on age. Patients could be
`under the age of 50 if they had a history of documented gastric or duodenal ulcer within the
`
`Page 9 of 25
`
`

`

`Division Director Review
`
`past 5 years. If patients were 50 years of age or older, they were not required to have a history
`of documented ulcer. Patients were randomized (1:1) between naproxen 500 mg twice daily or
`Vimovo (500 mg naproxen dose formulation) twice daily. Randomization was stratified based
`on low dose aspirin use. Endoscopies were performed at screening, one month, 3 months and
`6 months. Patients who discontinued prematurely returned for a final visit endoscopy.
`
`The primary endpoint was proportion of patients who developed gastric ulcer at any time
`throughout 6 months of treatment. An ulcer was defined as a mucosal break of at least 3 mm
`diameter with unequivocal crater depth. Sample size was determined based on an assumption
`that 15% of patients treated with naproxen would develop a gastric ulcer over the six months
`study, compared to 5% of patients treated with Vimovo. Key secondary endpoints included
`duodenal ulcers, expressed as a “tolerability” endpoint. In the submission, the Applicant
`proposed two efficacy endpoints that the Statistical reviewer concluded could only be viewed
`as exploratory: 1) incidence of gastroduodenal ulcers at any time throughout 6 months of
`treatment by low-dose aspirin use, and 2) incidence of gastroduodenal ulcers at any time
`throughout 6 months of treatment.
`
`The demographic analysis revealed that more than half of the population in both studies was
`female. The median age was 59 years. In Study 301, approximately 24% of patients used low
`dose aspirin and approximately 6% of patients had a history of ulcer. (Treatment arms were
`balanced). In Study 302, a similar proportion of patients were taking low dose aspirin (22% in
`the Vimovo arm and 24% in the naproxen arm), compared to Study 301; however a higher
`proportion had a history of ulcer (13% on the Vimovo arm and 9% on naproxen).
`
`The primary efficacy endpoint was analyzed using a CMH test stratified by use of low dose
`aspirin at randomization. The Applicant also conducted Kaplan-Meier estimates of the
`proportion of patients who developed ulcers. Time to documentation of gastric ulcer was
`calculated from the first day of study drug. Censoring occurred at the last day of endoscopic
`evaluation or date of withdrawal from the study if no gastric ulcer developed. The Applicant
`conducted sensitivity analyses classifying premature withdrawals without a confirmed gastric
`ulcer as having developed an ulcer at 6 months if they had discontinued from the trial due to a
`pre-specified upper gastrointestinal adverse event or if they developed a duodenal ulcer. The
`Applicant conducted additional exploratory analyses.
`
`FDA’s Statistical reviewer noted an imbalance in missing data between the studies and within
`Study 301. There was a higher premature discontinuation rate for Vimovo in Study 302 than
`301 (29% vs. 17%), and a higher premature discontinuation rate for naproxen in Study 301
`than in 302 (30.5% vs. 27.5%). The premature withdrawal rate was higher in the naproxen
`arm of Study 301 (30.5%) than the Vimovo arm (17%). She pointed out that missing efficacy
`data would conventionally be imputed as treatment failures for primary analyses; however, the
`higher missing data for the naproxen arm of Study 301 would significantly favor Vimovo in
`that approach. She concluded that the Applicant’s approach of imputing missing data as
`treatment success was the more conservative analysis from a regulatory perspective. The
`following table, which summarizes the primary efficacy results for the two studies, is
`reproduced from the Statistical review. Both studies resulted in a statistically significantly
`smaller proportion of patients who developed gastric ulcers over the 6 month period. The
`
`Page 10 of 25
`
`

`

`Division Director Review
`
`proportion of patients developing an ulcer in the naproxen arm exceeded the assumption upon
`which the sample size was based.
`
`
`
`The statistical reviewer recommended that the 3 key secondary endpoints proposed by the
`Applicant (two tolerability endpoints and the duodenal ulcer incidence endpoint) not be
`included in the efficacy section of the label because they were redefined late in the study. She
`had concern that the plan for statistical testing of the duodenal ulcer endpoint as a key endpoint
`was not included until late in the study.
`
`Exploratory analyses by gender, race, age, history of ulcer and use of low dose aspirin did not
`reveal internal inconsistencies for the primary endpoint.
`
`
`
` I
`
` concur with the reviewers’ conclusions that the Applicant has established that Vimovo is
`effective for reducing naproxen induced gastric ulcers. Esomeprazole carries the indication of
`reducing the risk of NSAID-induced gastric ulcers at doses of 20 mg/d and 40 mg/d (once
`daily dosing). The esomeprazole in Vimovo is an immediate release formulation that makes it
`vulnerable to degradation in gastric acid. The product is dosed twice daily. The single dose
`PK studies of Vimovo described in Section 5 of this review revealed that Vimovo was
`associated with lower esomeprazole exposures than Nexium 20mg. Repeat dose PK studies of
`Vimovo demonstrated higher exposures at Day 14. The two phase 3 trials demonstrated that
`immediate release esomeprazole in Vimovo is effective in this twice daily regimen for
`reducing naproxen induced gastric ulcers.
`
`Although Vimovo will be marketed in two dose levels of naproxen in the fixed combination,
`375 mg and 500 mg, the efficacy trials were only conducted at the 500 mg naproxen dose
`level. Esomeprazole efficacy in reducing gastric ulcers was demonstrated at the highest
`naproxen dose, so it is reasonable to assume that esomeprazole will contribute to the fixed
`combination at the lower naproxen 375 mg dose. A precedent for approving multiple dose
`levels of naproxen combined with a single dose level of a proton pump inhibitor can be found
`in the approval of the NDA for NapraPAC. The clinical review of that NDA indicates that
`patients in the clinical efficacy dataset were taking a range of naproxen doses, with the
`overwhelming majority taking 700-1000mg /day; however, the review does not indicate what
`
`Page 11 of 25
`
`

`

`Division Director Review
`
`proportion were taking 375 mg BID vs. 500 mg BID. Efficacy was not presented by dose
`level.
`
`Higher doses of NSAIDs are a risk factor for developing NSAID induced gastric ulcers;
`however, the Applicant submitted literature to support that naproxen 375 mg is associated with
`a risk for development of gastric ulcers. They cited a retrospective study (meeting abstract by
`Singh, Gurkirpal, et al.) of the MediCal database (California Medicaid program), that found
`that use of both naproxen 375 mg bid (750 mg/day) and naproxen 500 mg bid (1 g/day)
`significantly increases the risk of serious GI complications, defined as hospitalizations for
`complicated GU or DU (hemorrhage, perforation, or obstruction). The patients on 750 mg/day
`demonstrated a 2.95 increased risk ratio (RR) for hospitalization for complications of gastric
`and duodenal ulcers compared with controls (remote [>60 days previous] history of NSAID
`use). The corresponding RR for patients on naproxen 1 g/day was 3.13. This epidemiological
`analysis suggests that naproxen 375 mg twice daily is associated with GI risk, which in this
`study was slightly lower than that associated with the 500 mg twice a day. The Applicant also
`cited a short term (4 week) endoscopy trial (randomized, double blind) reported by Gomes, JA
`Melo, et al in Annals of the Rheumatic Diseases, 1993; 52: 881-885, in which 8.6% of patients
`treated with naproxen 375 mg BID developed gastroduodenal ulcers in that short time frame
`(6.9% developed gastric ulcers). A higher dose of naproxen was not studied in that trial. In
`comparison, the 1-month cumulative incidences of gastric ulcers in the EC-naproxen 500 mg
`arm of Study 301 and 302 were 13% and 10%, respectively.
`
`The proportion of patients who developed ulcers on the naproxen only comparator in the
`NapraPAC data set was higher than observed in the naproxen only arm of the two trials
`submitted this Vimovo NDA. This may be attributed to the risk of the population studied. In
`the NapraPAC study all patients had a history of gastric ulcer, compared to <15% of patients
`in the Vimovo trials. A lower dose of naproxen, 375 mg, may be associated with a lower
`proportion of patients developing gastric ulcers (which is supported by the information
`described above), which might result in a lower incremental improvement of risk for gastric
`ulcers with the addition of the esomeprazole magnesium in the fixed combination of Vimovo,
`particularly in patients who are not at high risk for developing ulcers; however, there is a
`substantial safety experience with esomeprazole and the risk benefit ratio supports approval in
`the lower dose naproxen combination. Should a definitive significant risk associated with
`esomeprazole use be identified in the future, the risk/benefit of this unstudied dose level may
`need to be re-examined.
`
`
`Osteoarthritis Trials
`Studies 307 and 309, reviewed by the Division of Anesthesia and Analgesia Products (DAAP),
`were double blind, randomized, placebo and active controlled trials of 12 weeks duration
`conducted to assess the efficacy of Vimovo for treatment of signs and symptoms of
`osteoarthritis (OA). The naproxen component of Vimovo has been approved as a single agent
`for treatment of signs and symptoms of osteoarthritis, rheumatoid arthritis, and ankylosing
`spondylitis. The trials were identical and included 3 arms: Vimovo (500mg/20 mg), placebo
`and celecoxib 200 mg once daily. Patients were randomized 2:2:1 (Vimovo:placebo:celcoxib).
`
`
`Page 12 of 25
`
`

`

`Division Director Review
`
`The planned primary efficacy analysis was a noninferiority analysis comparing Vimovo to
`celecoxib for the 3 primary endpoints: WOMAC pain subscale; WOMAC function subscale;
`and the Patient Global assessment (all 0-100 mm visual analogue scales). The DAAP clinical
`reviewers found all 3 efficacy endpoints acceptable. The Statistical reviewer noted that there
`was no plan for adjustment for multiplicity in the statistical plan but the Applicant had
`specified in the protocol that because noninferiority had to be demonstrated for all 3 primary
`endpoints, no adjustment was necessary.
`
`The protocol plan to use a noninferiority margin of 10 mm on each of the 0-100 VAS scales
`was not subject to agreement from FDA prior to study conduct, and the Applicant did not
`provide justification for selection of this margin. In a June 10, 2008, teleconference between
`the Applicant and FDA, the Applicant was advised that the FDA did not agree that a delta of
`10 mm for the noninferiority margin could be supported, and that the totality of evidence from
`the trials would be used to assess the Applicant’s desired comparison to celecoxib (since the
`trials contain a placebo arm). The clinical trials that established the efficacy of celecoxib did
`not utilize the same scales and the questions used to assess symptoms were worded differently.
`The Statistical reviewers and Clinical reviewers concluded that the celecoxib treatment effect
`relative to placebo (as a foundation for the noninferiority analysis) was best determined within
`Studies 307 and 309. Because the observed treatment effect of celecoxib vs. placebo in these
`trials was less than 10 mm (6-7 mm range across the endpoints in Study 307) and because
`celecoxib was not significantly different from placebo in Study 309 (with treatment effect
`sizes of 1-1.5 mm), the reviewers concluded that the 10 mm noninferiority margin proposed by
`the Applicant was not acceptable. The study results did not establish noninferiority of Vimovo
`to celecoxib
`
`
`
`
`
`The reviewers, however, determined that the comparison of Vimovo to placebo established its
`efficacy for the indication of treatment of signs and symptoms of OA. Vimovo was found to be
`statistically significantly different from placebo for all 3 efficacy endpoints. The statistical
`reviewer stated that the efficacy results were “consistent across the 3 endpoints and the
`alternative imputation methods.” I concur with the DAAP Clinical and Statistical reviewers
`that the results of these two studies provide sufficient evidence to support the efficacy of
`Vimovo (naproxen component) for the indication of treatment of signs and symptoms of
`osteoarthritis,
`
`
` Although the analysis comparing
`Vimovo to placebo in these trials was a prespecified secondary endpoint, one could argue that
`it technically should not have been tested after failure to establish noninferiority in the primary
`analysis. However, I agree that the bioequivalence of the naproxen component of Vimovo to
`the reference drug EC-Naprosyn, which is approved for the Applicant’s proposed indications
`for the naproxen component of Vimovo, supports the validity of the observed outcome of the
`comparison to placebo in these trials.
`
`The submitted trials only evaluated osteoarthritis and only evaluated the 500 mg naproxen
`dose level. The Applicant proposes that the indication for Vimovo include treatment of signs
`and symptoms of rheumatoid arthritis and ankylosing spondylitis, and they propose to market
`a Vimovo dose combination that includes 375 mg of naproxen. Naproxen is already approved
`
`Page 13 of 25
`
`(b) (4)
`
`(b) (4)
`
`

`

`Division Director Review
`
`and marketed as a single agent for treatment of signs and symptoms of osteoarthritis,
`rheumatoid arthritis and ankylosing spondylitis. Both of the proposed naproxen dose levels
`(375mg and 500 mg) are approved for treatment of these indications. The pharmacokinetic
`evaluation of Vimovo found no evidence of drug drug interaction between the naproxen and
`esomeprazole in Vimovo. The clinical pharmacology reviewers found that both doses of the
`referenced naproxen product, EC-Naprosyn, were bioequivalent to the respective Vimovo
`doses that contained the same amount of naproxen. Therefore there is adequate evidence that
`the naproxen component of Vimovo will contribute to the labeled treatment effect in all 3
`conditions and at both dose levels.
`
`In summary, I concur with the Clinical and Statistical reviewers of this application that the
`Applicant has established the efficacy of each component of this fixed combination product.
`
`
`8. Safety
`This is a 505(b)(2) application. The two product components of Vimovo, naproxen and
`esomeprazole, have been marketed in the US since 1976 and 2001, respectively. The Clinical
`pharmacology reviewers found that the naproxen component is bioequivalent to the referenced
`approved naproxen product, EC-Naprosyn, and determined in a single dose study that the
`AUC of the 20 mg immediate release eso